1. Field of the Invention
This invention relates to the design of fluorescent compounds used as glucose sensors that have selected fluorescent properties, such as long fluorescence lifetimes, long absorption or emission wavelengths, or high quantum yields. More particularly, the invention relates to biomedical sensors for continuous transdermal optical transduction of tissue glucose concentration for the treatment of diabetes.
2. Description of Related Art
Molecular recognition coupled with fluorescent quenching is a promising analytical technique for optical transduction of analyte concentrations in bioassays. Several schemes employing fluorescent resonant energy transfer (FRET) have been proposed. Intramolecular electron transfer schemes, also called photoinduced electron transfer or PET, appear superior from the standpoint of chemical robustness and simplicity.
U.S. Pat. No. 5,503,770 to James et al. discloses a fluorescent compound used for the detection of saccharides or sugars such as glucose. The use of these fluorescent compounds was extended in U.S. Pat. No. 6,002,954 to Van Antwerp et al., by incorporating the compounds in an implantable optical sensor for transduction of glucose concentration for the treatment of diabetes. The fluorescent transducer is implanted 1-3 mm below the surface of the skin and optically interrogated externally to determine the level of tissue glucose in diabetic patients. A minimally-invasive, continuous glucose sensor is of great benefit to patients in achieving tighter blood-glucose control when combined with existing insulin pumps.
One of the systems described in U.S. Pat. No. 6,002,954 is made of anthracene boronate, which is excited at about 380 nm and emits at about 420 nm. Unfortunately, excitation and fluorescent emission at wavelengths in the range of about 360-420 nm is only weakly transmitted through skin. The light transmission through the skin must be significantly increased for the molecule to be useful as a sensor over the clinically relevant range of tissue glucose concentration. In addition, it may also be advantageous to modify other properties of the system associated with the fluorescent portion of the sensing molecule. Such properties may include the quantum yield, the fluorescence lifetime, photostability, chemical stability, and biocompatibility.
To improve transmission of the signal through the skin, the fluorescent compound should operate at longer wavelengths than about 450 nanometers. The transmission through a few millimeters of skin increases logarithmically with wavelengthxe2x80x94from 0.1% at about 400 nm to almost 100% at 850 nm. Thus, the longer the wavelength, the greater the transmission through skin. An excitation and emission wavelength greater than about 600-650 nm is an enormous improvement over about 400-450 nm. Because of the significant increase in optical skin transmission at longer wavelengths, a practical glucose sensor can operate more efficiently, more accurately, and with a greater signal-to-noise ratio.
In addition, it is advantageous to match the peak excitation wavelength with an existing light source (such as an LED or diode laser). Furthermore, by operating at longer wavelengths, there is a reduction in the tissue autofluorescence background. Further progress in creating such compounds, however, entails the formidable task of synthetically assembling various combinations of fluorophores and glucose-recognizers into integrated molecules that have the desired fluorescent properties, such as operating at longer wavelengths (450-700 nm), while simultaneously retaining the requisite glucose transduction properties.
This invention addresses the optical transmission problem and provides exemplary fluorescent compounds that have been demonstrated to exhibit the needed photochemical behavior and operate in a wavelength range that makes a subcutaneous fluorescent glucose sensor practical.
The invention disclosed herein provides fluorescent analyte binding compounds defined by a specified formula and wherein substituent molecules of these compounds are selected to have complimentary molecular properties which can be determined via simplified version of the Rehm-Weller equation. The representative molecules of the invention have a number of exceptional activities that make them uniquely suited for incorporation into a sensor for analytes such as sugars. These activities include highly desirable fluorescent properties such as longer excitation and emission wavelengths, properties which are highly compatible with their use in minimally implantable systems for the continuous transdermal monitoring of blood glucose concentrations.
As shown by the disclosure provided herein, it is an object of the present invention to provide fluorescent compounds that may be incorporated into a sensor for saccharides or sugars and that have desired fluorescent properties, such as longer absorption or emission wavelengths, high quantum yields, or long fluorescence lifetimes. It is an object of the present invention to provide glucose sensing fluorescent molecules that operate at emission wavelengths greater than about 450 nm. It is another object of the present invention to provide glucose sensing fluorescent molecules that operate at excitation wavelengths greater than about 400 nm. The sensors can then be used to sense glucose concentrations in media with high visible blue and ultraviolet opacities such as human tissue.
Another object of the invention is to provide a sensor comprising a biocompatible polymer containing covalently bonded fluorescent molecules that specifically and reversibly bind glucose. Upon binding with the glucose, the fluorescence quantum yield of the molecules is increased, resulting in increased fluorescent emission at increasing glucose concentrations. Glucose transduction is achieved at wavelengths that allow the sensor to be implanted subcutaneously and permit improved optical transmission through skin.
The invention basically involves a fluorescent compound having three functional components in one molecule: a substrate-recognition component (typically a substituted aryl boronic acid), a fluorescence xe2x80x9cswitchxe2x80x9d that is mediated by a substrate recognition event (typically an amine), and a fluorophore. The sensor molecule is designed so that the photo-excited fluorophore and the boron atom compete for the unbonded amine electrons. In the absence of glucose binding, electron transfer occurs predominantly with the fluorophore, causing fluorescent quenching and subsequently weak emission. When glucose is bound to the boronate group, the average charge on the boron atom becomes more positive, which increases the attraction of the unbonded electrons, preventing electron transfer, thus disabling the fluorescent quenching, and therefore causing strong emission.
Achieving successful interplay of these molecular components, resulting in optochemical glucose sensitivity, requires a precise matching of their electrochemical and photophysical properties. For example, given a functioning glucose-sensor molecule, shifting the wavelength of operation cannot be accomplished by simply substituting the original fluorophore with another fluorophore of the desired wavelength. Rather, a fluorophore and a switch with compatible reduction and oxidation potentials and photo-excited state energy must be selected. Similarly, other modifications of the molecule, such as to accommodate immobilization of the molecule in a biocompatible substrate (suitable for surgical implantation), could alter the oxidation potential of the xe2x80x98switchxe2x80x99 component, thereby upsetting the electrochemical balance required for operation. Therefore, modifications, variations or substitutions of these three components must be evaluated both electrochemically and optochemically so that compatibility can be established prior to the often lengthy process of synthetically incorporating them into a new molecular system. Experimental measurements were carried out for the purpose of screening prospective components and for evaluating the opto-chemical glucose sensitivity of compounds synthetically derived from those components.
When these components are assembled into the final sensor molecule, interactions between the assembled components are easily capable of modifying both the electrochemical potentials and photophysical characteristics of the stand-alone components. Surprisingly, it was found that for the boronate glucose sensing molecule having the specific formula described in this application, these interactions are minimal. Therefore, measurements made on the individual components such as the specific fluorophore afford an accurate prediction of the PET behavior of the final assembled molecule. Consequently, a simplified version of the Rehm-Weller equation can be employed to identify saccharide binding fluorescent compounds having the desired activities so that they can be synthesized for a variety of applications.
Glucose transduction has been successfully achieved by three typical classes of candidate molecules that were investigated: transition metal-ligand boronate compounds and conjugated organic heterocyclic ring system compounds that are oxazine, oxazine-one, oxazone, and thiazine boronate compounds and anthracene boronate compounds. Typically, these compounds are excited at wavelengths greater than about 400 nm. In addition, these compounds typically operate (emit) at wavelengths greater than about 450 nm.
The present invention is thus useful in the detection, localization, and quantification of saccharide or sugar concentration in media that are opaque in the visible blue or ultraviolet spectral regions. A glucose sensor according to the present invention can be incorporated into a minimally invasive, implantable system for continuous transdermal monitoring of glucose levels in diabetic patients. In addition, glucose sensors can be designed having other desired molecular properties, such as a longer fluorescence lifetime, specific excitation and emission wavelengths, high quantum yields, photostability, chemical stability, high water solubility, low temperature sensitivity, or low pH sensitivity.